A method, an apparatus, and a computer program, which can suppress a low frequency range component with a small amount of calculation, and can achieve a noise suppression of high quality, are provided. The noise superposed in a desired signal of an input signal is suppressed by converting the input signal to a frequency domain signal; correcting an amplitude of the frequency domain signal to obtain an amplitude corrected signal; obtaining an estimated noise by using the amplitude corrected signal; determining a suppression coefficient by using the estimated noise and the amplitude corrected signal; and weighting the amplitude corrected signal with the suppression coefficient.
Legal claims defining the scope of protection. Each claim is shown in both the original legal language and a plain English translation.
1. A noise suppressing method for suppressing noise included in an input signal, comprising: eliminating an offset of the input signal to obtain an offset eliminated signal; converting the offset eliminated signal to a frequency domain signal; correcting an amplitude of the frequency domain signal to obtain an amplitude corrected signal; obtaining an estimated noise by using the amplitude corrected signal; determining a suppression coefficient by using the estimated noise and the amplitude corrected signal; and weighting the amplitude corrected signal with the suppression coefficient.
A noise reduction method suppresses noise within an audio signal. First, it removes any DC offset from the input signal. This offset-corrected signal is then transformed into the frequency domain. The amplitude of this frequency-domain signal is adjusted. An estimate of the noise present in the signal is then calculated using this amplitude-adjusted signal. A suppression coefficient is determined based on both the noise estimate and the amplitude-adjusted signal. Finally, the amplitude-adjusted signal is weighted using this suppression coefficient to reduce the noise.
2. The noise suppressing method according to claim 1 , wherein the correction is to correct the amplitude of the frequency domain signal to include a desired high-pass characteristic along with the offset eliminating process.
The noise reduction method from the previous description, which suppresses noise within an audio signal by removing DC offset, converting to frequency domain, correcting amplitude, estimating noise, determining a suppression coefficient, and weighting the amplitude-adjusted signal, also incorporates a high-pass filter during the amplitude correction. This high-pass filtering is done in conjunction with the DC offset removal step. This combined step adjusts the amplitude of the frequency domain signal to both remove the DC offset and filter out low frequency components.
3. The noise suppressing method according to claim 2 , wherein the desired high-pass characteristic suppresses a component close to a direct current, and passes a voice.
Building upon the noise reduction method described previously, which suppresses noise by removing DC offset, converting to the frequency domain, correcting amplitude with a high-pass filter, estimating noise, determining a suppression coefficient, and weighting the amplitude-adjusted signal, the high-pass filter is specifically designed to attenuate frequencies close to DC while allowing voice frequencies to pass through relatively unchanged. This ensures that the noise suppression focuses on very low-frequency noise without significantly affecting speech content.
4. The noise suppressing method according to claim 1 , further comprising: correcting a phase of the frequency domain signal to obtain a phase corrected signal; and converting a result that is obtained by weighting the amplitude corrected signal with the suppression coefficient and the phase corrected signal to a time domain signal.
The noise reduction method described earlier, which suppresses noise by removing DC offset, converting to the frequency domain, correcting amplitude, estimating noise, determining a suppression coefficient, and weighting the amplitude-adjusted signal, is further enhanced by processing the phase of the frequency domain signal. A phase correction is applied to generate a phase-corrected signal. After weighting the amplitude-corrected signal with the suppression coefficient, this result is combined with the phase-corrected signal. Finally, this combined signal is transformed back into the time domain, resulting in a noise-reduced audio signal.
5. A filtering method for suppressing a specific frequency component of an input signal, comprising: executing a first filtering process for an input signal in a time domain to obtain a time domain filtered signal; converting the time domain filtered signal to a frequency domain signal for each frame configured with a plurality of samples; and executing a second filtering process for the frequency domain signal in a frequency domain to obtain a frequency domain filtered signal, wherein the first filtering process suppresses at least a direct current component.
A filtering method suppresses specific frequencies in an audio signal. It begins by applying a first filter in the time domain to the input signal. The output of this filter is then divided into frames, each containing multiple samples. These time-domain frames are then converted into the frequency domain. A second filter is then applied to the frequency-domain signal. Critically, the first filter suppresses at least the DC component of the input signal. This pre-filtering in the time domain is a key part of the overall frequency suppression process.
6. The filtering method according to claim 5 , wherein a characteristic obtained by combining the first filtering process and the second filtering process suppresses a component close to a direct current, and passes a voice.
Combining the filtering methods, which suppress specific frequencies by first filtering in the time domain to reduce DC components, then converting to the frequency domain and applying a second filter, the overall filtering characteristic attenuates frequencies close to DC while passing voice frequencies. The combined effect of the time-domain and frequency-domain filters creates a filter that primarily targets very low frequencies, including DC, while minimizing the impact on speech content in the audio signal.
Cooperative Patent Classification codes for this invention. Click any code to explore related patents in that topic.
June 25, 2012
July 16, 2013
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.